In May of 2008, The Endocrine Society issued a Clinical Guideline on the diagnosis of Cushing’s syndrome (1). The purpose of the guideline was to provide recommendations for testing based on the medical literature and a consensus of experts. Another goal was to enhance early detection of the disorder and help reduce false diagnoses. An international task force reviewed the literature to address issues of the benefit of testing, who should be tested, which tests are optimal, and the pitfalls of the tests and their interpretation. A statistical evaluation of the medical literature regarding the usefulness of each individual test was compiled to determine the reliability of each test. The statistical evaluation was used by the experts to formulate the Guideline. The Guideline was approved by the Endocrine Society’s Clinical Guidelines Subcommittee, Clinical Affairs Core Committee and the Society’s Council after members responded to a request for their review and comments.
The risks and benefits of testing for Cushing’s syndrome
The Guidelines recognize that detection of Cushing’s syndrome is extremely important because successful treatment can reduce or eliminate the risk of bad medical outcomes associated with the condition. These outcomes, including increased clotting, (which can lead to heart attack, stroke or pulmonary embolism), hypertension, diabetes, infection, osteoporosis, etc., occur most often in patients with moderate to severe over-production of cortisol. While severe cases of Cushing’s are usually easy to identify clinically and have clearly abnormal test results, many patients present to their physicians with a spectrum of clinical symptoms and variable levels of hypercortisolism. Mild cases can be challenging to diagnose and can progress with additional symptoms associated with further rises in cortisol. While treatment is highly desirable prior to the onset of severe symptoms, it can be challenging to accurately identify those with mild Cushing’s who will progress over time, versus those patients that may have an abnormal test result, but do not have Cushing’s syndrome. The task force noted that data regarding the benefit of treatment in patients with very mild over-production of cortisol are limited and conflicting and treatment did not always result in improvement of clinical symptoms. Thus, patients with few clinical and laboratory abnormalities have a high risk of receiving an incorrect diagnosis of Cushing’s syndrome resulting in invasive testing and surgical treatment that may be harmful, and without benefit. Because mild Cushing’s syndrome generally progresses, it is possible that early detection and treatment would result in better patient outcome, however, there currently are no data to support this. Thus, while extremely early detection and treatment may not be critical, the disorder needs to be treated before symptoms become severe.
Who should be tested for Cushing’s syndrome?
Although Cushing’s syndrome is rare, its symptoms, including obesity or weight gain, hypertension, extra hair growth, menstruation problems, depression, and tiredness, are common in the general population. The task force suggested that testing be done in the following cases:
- Patients with multiple clinical symptoms, particularly those more indicative of Cushing’s such as easy bruising, flushed face, muscle weakness in the hips and arms, reddish, purple striae (stretch marks) and those patients whose clinical symptoms have progressed over time.
- Patients who develop symptoms not typical for their age, such as hypertension and osteoporosis.
- Children who have an increase in weight and a decrease in growth.
- Patients who have an accidentally discovered adrenal tumor.
Cushing’s syndrome patients may have a number of symptoms common in the population such as hypertension, diabetes and weight gain. The Guidelines encourage physicians to consider Cushing’s syndrome as a cause of these symptoms, particularly if other signs of Cushing’s syndrome are present. At this time, widespread testing is not recommended in at risk populations such as diabetics who do not have other features of the disorder. Past photographs illustrating a change in the patient’s appearance can be useful to physicians.
Testing in these groups who are most likely to have Cushing’s syndrome should reduce the chance of false diagnoses in patients with few typical characteristics. Conditions that are associated with high cortisol and some clinical features of Cushing’s syndrome, but are not actually Cushing’s syndrome include pregnancy, depression and other psychiatric conditions, alcohol dependence, glucocorticoid resistance, extreme obesity, and poorly controlled diabetes mellitus.
Before testing, patients with the clinical features of Cushing’s syndrome should be asked about whether they have received cortisol-like medications, including injections, inhaled medication, or skin preparations, as well as herbal or over the counter formulations. These agents can cause Cushing’s syndrome, and may lead to a false abnormal test, and these patients will be “cured” when these items are no longer used.
How to test for Cushing’s syndrome
The guidelines recommend that one of four tests be used for the first evaluation:
- 24-hour urine cortisol measurements (at least 2 measurements),
- Late-night salivary cortisol measurements (at least 2 measurements),
- The overnight 1 mg dexamethasone suppression test, or
- The two-day 2 mg dexamethasone suppression test (0.5 mg every 6 hours for 8 doses).
Patients with two abnormal results would undergo testing for the cause of Cushing’s syndrome, if there is no reason to doubt the validity of the results. Those with two normal results would not, in general, undergo further testing, according to the recommendations. An exception would be patients in whom cyclic Cushing’s syndrome is suspected, who would have repeat testing. Repeat testing also would be done in patients who develop new features of Cushing’s syndrome over time and in those with mixed normal and abnormal results after the initial evaluation.
Problems with the choice and interpretation of tests
The task force identified a number of problems with the tests and their interpretation. All of the tests require laboratory measurements, whose accuracy may vary from one laboratory to another. Additionally, the tests can give a false result in a variety of situations. For example, shift workers may not have a (normal) low late night cortisol value, so using late night blood or saliva cortisol testing may suggest, falsely, that they have Cushing’s syndrome. Also, a 24-hour UFC is the test of choice during pregnancy, but needs to be interpreted with knowledge that the “normal” range is increased in later pregnancy. Additionally, many medications affect the way that dexamethasone is cleared from the body, and may cause a false result. Similarly, some medications can falsely raise cortisol values. Practitioners should be aware of these pitfalls and individualize the choice of tests for each patient in light of these issues. The panel did observe that “referral to endocrinology centers with expertise and interest in Cushing’s syndrome in patients with abnormal initial testing is likely to be associated with better patient outcomes.”
Because Cushing’s syndrome is potentially fatal if untreated, patients most likely to have the condition should be tested and treated. To minimize false results the testing strategy should be individualized for each patient. When interpreting the results, pitfalls of the tests and possible laboratory error should be considered.
Author: Lynnette Nieman, MD (Summer, 2008)
Editor’s Note: This article is condensed from information contained in the full Clinical Guideline.
Dr. Nieman is a Senior Investigator in the Intramural Research Program on Reproductive and Adult Endocrinology at The Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) and Associate Director, Intramural Endocrinology Training Program at the National Institutes of Health in Bethesda, MD. Dr. Nieman has many years of experience in the diagnosis and treatment of Cushing’s and conducts research in the area.